Positron emission tomography (PET) has been widely accepted for clinical imaging for oncologic, cardiac and neurological applications. The current trend of design favors high-efficiency median-resolution scanners to minimize the scan time for whole body imaging protocols. The smallest object detectable and the quantitative accuracy of PET imaging are directly related to the scanner resolution. The emerging field of molecular imaging to study disease models, pharmacokinetics of new drugs, and gene expression using small laboratory animals via in vivo imaging has shown promising results. The techniques and models developed from laboratory animal studies will soon require clinical trials to provide validation and further development for human applications. The success of such translation relies on high-resolution human PET systems to advance the molecular imaging research to clinical applications. The goal of this project is to develop the technology required to provide high-resolution PET images of the organs of interest at a fraction of the cost of a dedicated high-resolution PET scanner. The proposed design is a detector insert that can be attached to a whole-body PET scanner to provide high-resolution images within a smaller field of view, a concept similar to the use of surface coils in MRI to improve its image resolution. Since the device is designed as an accessory detector, its cost is significantly lower than a complete system. The image resolution within the regions of interest can be as low as 3 mm FWHM or smaller, depending on the design of the detectors, the dimension of the insert, and the original system resolution. The sensitivity of such a device may be more than twofold of a stand-alone high resolution system using the same detectors. The R21 phase of this project provides feasibility study of the design, while the R33 phase is to implement a prototype device for an existing PET scanner. The specific aims of the R21 phase are (1) develop APD-based high-resolution PET detector, (2) validate the image resolution and sensitivity of the proposed device via simulation and experiment, and (3) evaluate the effect of scatter from the device and develop a rejection technique. The specific aims of the R33 phase are: (1) Construct the device, (2) integrate the device to an existing PET scanner, (3) develop image reconstruction algorithms for the special geometry, and (4) test the device and evaluate its performance. The immediate applications of the prototype device include high-resolution imaging for head and neck tumors and lymph nodes, brain tumors, and brain functions. The design earl be extended to develop special inserts to image cancers involving breasts, lungs, and prostate. The outcome of this project will make a significant advancement in high resolution human PET imaging for both clinical and research applications.